deleted by creator

As @[email protected] pointed out, the author considers something as small as spawning a separate process for each window to mean a “non-native experience” (wait till they see how web browsers work)

To be fair, giving a company that’s been failing to get themed icons to work on Android for almost four years now less than a month to make a significant change to a core part of their software is… quite weird?

Like, the EU usually gives companies at least half a year to comply with smaller demands than this, because companies with such a huge bureaucracy load wouldn’t even be able to change an app logo in such a short amount of time.

Yeah, this seems properly configured. No clue why it isn’t working for you.

The only app that doesn’t auto-update for me is Fdroid itself (ironically), because it targets an old Android version. Running Android 14 on a Pixel, so with the strongest Google fuckery.

Are you sure your Fdroid client is up to date? The new API was implemented in 1.19, and apparently I even misremembered and all you have to do to enable Fdroid to auto update its apps is to manually update them for one last time (so no fresh installation required).

Another long shot: there’s an option to force the old installation method hidden in expert settings - maybe you could check if that isn’t enabled?

On a normal unmodified phone you have to manually confirm each app you want to install. so no auto-updates in the background etc.

Background app updates are possible since Android 12, Fdroid just took two years to implement the new API (and you have to do a fresh install of the apps - apps already installed using the old API still require confirmation on each update). There is still friction on the initial install though.

Both? It’s pretty well explained in the rest of the text (you don’t even have to click a link)

It was up to the Commission, which has exclusive powers to set the bloc’s commercial policy, to break the gridlock and ensure the duties go through.

The European Commission made the decision after the member countries failed to agree on how to proceed.

What error? It gave you a string of tokens that seemed likely according to its training data. That’s all it does.

If you ask it what color is the sky, it will tell you it’s blue not because it knows that’s true, but because these words “fit together”. Pretty much the only way to avoid this issue is to put some kind of filter in front of the LLM which will try to catch prompts that are known to produce unwanted results, and silently replace your prompt with something like “say: sorry, I don’t know”.

I’m being very reductive here, but that’s the principle of how these things work - the LLMs are not capable of determining the truthfulness of their responses.

proprietary Google-only format

Keyhole Markup Language

KML became an international standard of the Open Geospatial Consortium in 2008.

(…)

The KML 2.2 specification was submitted to the Open Geospatial Consortium to assure its status as an open standard for all geobrowsers.

That really depends on the technology used. For example, all modern Ethernet standards (which includes both copper and fiber optic) are full duplex, meaning they can provide the full bandwidth in both directions at once. So a gigabit Ethernet link can do a gigabit in one direction AND a gigabit in the other direction at the same time (but not two gigabits in one direction).

In my very limited experience with my 5400rpm SMR WD disk, it’s perfectly capable of writing at over 100 MB/s until its cache runs out, then it pretty much dies until it has time to properly write the data, rinse and repeat.

40 MB/s sustained is weird (but maybe it’s just a different firmware? I think my disk was able to actually sustain 60 MB/s for a few hours when I limited the write speed, 40 could be a conservative setting that doesn’t even slowly fill the cache)

Your mileage may vary - your experience might be different for one reason or another

Vista’s problem was just the terrible third party drivers and the fact that it was preinstalled on machines it had no business running on. 7 didn’t improve much on it (except fixing the UAC prompt so that it no longer made you feel like you’re using Linux with misconfigured sudo timeout), but it had the benefit of already having working drivers from Vista and proper hardware capable of running Vista/7.

Zig didn’t come to my mind when I was writing my comment and I agree that it’s probably a decent option (the only issue I can think of is its somewhat small community, but that’s not a technical issue with the language).

My argument against Go and Java is garbage collection - even if Java’s infamous GC pause can apparently be worked around with a specialized JVM, I’m pretty sure it still comes at the cost of higher memory usage and wasted CPU cycles compared to some kind of reference counting or Rust’s ownership mechanism (not sure about the proper term for that). And higher memory usage is definitely not something I want to see in my browser, they’re hungry enough as is.

Why not just say Rust? There isn’t really anything else that would provide good enough performance for a browser engine with modern heavy webpages while also fixing some major pain point of C/C++

They’re not doing a recall, but that doesn’t mean they won’t somehow compensate big OEMs for their warranty issues.

Right, now get a borderline computer-illiterate person to connect to your network, ensure their firewall isn’t misconfigured to block all incoming traffic (with TeamViewer, this configuration would still work because the device just connects to the TV server) and open and set up a completely separate screen sharing program.

I know none of these steps are difficult if you have any idea what you’re doing, but I’ve met plenty of people who would most likely need assistance going through the motions. Funnily enough, the best way to do it remotely would probably be to get them to install TeamViewer to then set this up for them remotely.

By the way, as far as networking goes, Tailscale does the same thing TeamViewer does, just for a VPN instead of a screen sharing application - it will try to do all the NAT punchthrough techniques and IPv6 connection and fall back on tunneling through relay servers if all else fails. It’s not any more of a direct connection than TV.

Convenience (after you install it, all you have to do is enter the code and you’re connected, no other setup required), familiarity (it’s the default name people will think of or find if they want remote access - that alone means they can get away with pushing their users slightly more) and - IMHO most importantly - connectivity: if two computers can connect to the TeamViewer servers, they will be able to connect to each other.

That’s huge in the world of broken Internet where peer to peer networking feels like rocket science - pretty much every consumer device will be sitting behind a NAT, which means “just connecting” is not possible. You can set up port forwarding (either manually or automatically using UPnP, which is its own bag of problems), or you can use IPv6 (which appears to be currently available to roughly 40% users globally; to use it, both sides need to have functional IPv6), or you can try various NAT traversal techniques (which only work with certain kinds of NAT and always require a coordinating server to pull off - this is one of the functions provided by TeamViewer servers). Oh, and if you’re behind CGNAT (a kind of NAT used by internet providers; apparently it’s moderately common), then neither port forwarding or NAT traversal are possible. So if both sides are behind CGNAT and at least one doesn’t have IPv6, establishing a direct link is impossible.

With a relay server (like TeamViewer provides), you don’t have to worry about being unable to connect - it will try to get you a direct link, but if that fails, it will just act as a tunnel and pass the data between both devices.

Sure, you can self host all this, but that takes time and effort to do right. And if your ISP happens to use CGNAT, that means renting a VPS because you can’t host it at home. With TeamViewer, you’re paying for someone else to worry about all that (and pay for the servers that coordinate NAT traversal and relay data, and their internet bandwidth, neither of which is free).

If it doesn’t come at the expense of battery wear, then sure, lower charge time is just better. But that would make phone batteries the only batteries that don’t get excessively stressed when fast charging. Yeah, phone manufacturers generally claim that fast charging is perfectly fine for the battery, but I’m not sure I believe them too much when battery degradation is one of the main reasons people buy new phones.

I have no clue how other manufacturers do it (so for all I know they could all be doing it right and actually use slow charging), but Google has a terrible implementation of battery conservation - Pixels just fast charge to 80%, then wait until some specific time before the alarm, then fast charge the rest. Compare that to a crappy Lenovo IdeaPad laptop I have that has a battery conservation feature that sets a charge limit AND a power limit (60% with 25W charging), because it wouldn’t make sense to limit the charge and still use full 65W for charging.

It doesn’t slow charge, at least not on Pixel 7a. Well, you could argue whether 20W is slow charging, but it’s all this phone can do.

It just charges normally to 80%, stops, and then resumes charging about an hour or two before the alarm. And last time I used it, it had a cool bug where if it fails to reach 80% by the point in time when it’s supposed to resume charging, it will just stop charging no matter what the current charge level is. Since that experience, I just turned this feature off and charge it in whenever it starts running low.